From bd907f4c44f9424a9c5817b38bd24e68193d222e Mon Sep 17 00:00:00 2001
From: TrellixVulnTeam <charles.mcfarland@trellix.com>
Date: Wed, 19 Oct 2022 18:57:10 +0000
Subject: [PATCH] Adding tarfile member sanitization to extractall()

---
 .../official/nlp/transformer/data_download.py | 21 ++++++++++++++++++-
 .../research/deep_speech/data/download.py     | 21 ++++++++++++++++++-
 .../models/research/neural_gpu/wmt_utils.py   | 21 ++++++++++++++++++-
 3 files changed, 60 insertions(+), 3 deletions(-)

diff --git a/mobile_deployment/tensorflow/slim/models/official/nlp/transformer/data_download.py b/mobile_deployment/tensorflow/slim/models/official/nlp/transformer/data_download.py
index e5f6668..9cf11b1 100644
--- a/mobile_deployment/tensorflow/slim/models/official/nlp/transformer/data_download.py
+++ b/mobile_deployment/tensorflow/slim/models/official/nlp/transformer/data_download.py
@@ -208,7 +208,26 @@ def download_and_extract(path, url, input_filename, target_filename):
   # Extract compressed files
   logging.info("Extracting %s." % compressed_file)
   with tarfile.open(compressed_file, "r:gz") as corpus_tar:
-    corpus_tar.extractall(path)
+    def is_within_directory(directory, target):
+        
+        abs_directory = os.path.abspath(directory)
+        abs_target = os.path.abspath(target)
+    
+        prefix = os.path.commonprefix([abs_directory, abs_target])
+        
+        return prefix == abs_directory
+    
+    def safe_extract(tar, path=".", members=None, *, numeric_owner=False):
+    
+        for member in tar.getmembers():
+            member_path = os.path.join(path, member.name)
+            if not is_within_directory(path, member_path):
+                raise Exception("Attempted Path Traversal in Tar File")
+    
+        tar.extractall(path, members, numeric_owner=numeric_owner) 
+        
+    
+    safe_extract(corpus_tar, path)
 
   # Return file paths of the requested files.
   input_file = find_file(path, input_filename)
diff --git a/mobile_deployment/tensorflow/slim/models/research/deep_speech/data/download.py b/mobile_deployment/tensorflow/slim/models/research/deep_speech/data/download.py
index 5ded037..020dd9f 100644
--- a/mobile_deployment/tensorflow/slim/models/research/deep_speech/data/download.py
+++ b/mobile_deployment/tensorflow/slim/models/research/deep_speech/data/download.py
@@ -78,7 +78,26 @@ def _progress(count, block_size, total_size):
     tf.logging.info(
         "Successfully downloaded %s, size(bytes): %d" % (url, statinfo.st_size))
     with tarfile.open(tar_filepath, "r") as tar:
-      tar.extractall(directory)
+      def is_within_directory(directory, target):
+          
+          abs_directory = os.path.abspath(directory)
+          abs_target = os.path.abspath(target)
+      
+          prefix = os.path.commonprefix([abs_directory, abs_target])
+          
+          return prefix == abs_directory
+      
+      def safe_extract(tar, path=".", members=None, *, numeric_owner=False):
+      
+          for member in tar.getmembers():
+              member_path = os.path.join(path, member.name)
+              if not is_within_directory(path, member_path):
+                  raise Exception("Attempted Path Traversal in Tar File")
+      
+          tar.extractall(path, members, numeric_owner=numeric_owner) 
+          
+      
+      safe_extract(tar, directory)
   finally:
     tf.gfile.Remove(tar_filepath)
 
diff --git a/mobile_deployment/tensorflow/slim/models/research/neural_gpu/wmt_utils.py b/mobile_deployment/tensorflow/slim/models/research/neural_gpu/wmt_utils.py
index ef83191..dd50561 100644
--- a/mobile_deployment/tensorflow/slim/models/research/neural_gpu/wmt_utils.py
+++ b/mobile_deployment/tensorflow/slim/models/research/neural_gpu/wmt_utils.py
@@ -84,7 +84,26 @@ def get_wmt_enfr_train_set(directory):
                                  _WMT_ENFR_TRAIN_URL)
     print("Extracting tar file %s" % corpus_file)
     with tarfile.open(corpus_file, "r") as corpus_tar:
-      corpus_tar.extractall(directory)
+      def is_within_directory(directory, target):
+          
+          abs_directory = os.path.abspath(directory)
+          abs_target = os.path.abspath(target)
+      
+          prefix = os.path.commonprefix([abs_directory, abs_target])
+          
+          return prefix == abs_directory
+      
+      def safe_extract(tar, path=".", members=None, *, numeric_owner=False):
+      
+          for member in tar.getmembers():
+              member_path = os.path.join(path, member.name)
+              if not is_within_directory(path, member_path):
+                  raise Exception("Attempted Path Traversal in Tar File")
+      
+          tar.extractall(path, members, numeric_owner=numeric_owner) 
+          
+      
+      safe_extract(corpus_tar, directory)
     gunzip_file(train_path + ".fr.gz", train_path + ".fr")
     gunzip_file(train_path + ".en.gz", train_path + ".en")
   return train_path